Other · model
Agents-A1-4B benchmark scores
As of 2026-11-22, Agents-A1-4B (Other) has measured scores on 8 benchmarks and estimated scores on 39 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-SciCode | 30.5% | estimated ± 3.0 pp, high confidence |
| GDPval-AA | 26.2% | estimated ± 7.8 pp, medium confidence |
| IFBench | 69.1% | measured |
| AA-LCR | 55.7% | estimated ± 12.5 pp, low confidence |
| LiveCodeBench v6 | 59.6% | measured |
| SciCode | 29.6% | measured |
| BrowseComp | 66.8% | measured |
| HLE w/ tools | 29.5% | estimated ± 5.5 pp, low confidence |
| IFEval | 94.8% | measured |
| LongBench v2 | 52.1% | measured |
| VITA-Bench | 40.3% | measured |
| τ²-bench results | 78.2% | measured |
| AA Coding Index | 11.8% | estimated ± 5.7 pp, medium confidence |
| APEX-Agents-AA | 14.3% | estimated ± 4.4 pp, high confidence |
| DeepSearchQA | 78.8% | estimated ± 6.8 pp, low confidence |
| SWE-bench Verified | 47.5% | estimated ± 5.6 pp, medium confidence |
| BFCL v4 | 60.4% | estimated ± 8.5 pp, low confidence |
| CyberGym | 53.9% | estimated ± 7.3 pp, medium confidence |
| JobBench | 25.2% | estimated ± 10.7 pp, low confidence |
| SkillsBench | 38.6% | estimated ± 4.1 pp, low confidence |
| SWE-bench Pro | 0.0% | estimated ± 6.0 pp, low confidence |
| WideResearch | 70.9% | estimated ± 3.5 pp, high confidence |
| AA Agentic Index | 22.0% | estimated ± 8.2 pp, medium confidence |
| AA-IFBench | 56.7% | estimated ± 7.4 pp, medium confidence |
| Vals LiveCodeBench | 81.3% | estimated ± 0.8 pp, low confidence |
| Gert Labs | 30.9% | estimated ± 11.1 pp, low confidence |
| AA Harvey LAB v1.0 | 87.2% | estimated ± 9.8 pp, low confidence |
| AA Tau3 Banking | 26.8% | estimated ± 3.6 pp, low confidence |
| MazeBench | 0.0% | estimated ± 6.4 pp, low confidence |
| OSWorld-Verified | 59.4% | estimated ± 5.7 pp, medium confidence |
| Vals SWE-bench | 31.6% | estimated ± 5.3 pp, low confidence |
| CWE-Bench-AA | 44.1% | estimated ± 6.4 pp, medium confidence |
| AA Cyber Index | 33.2% | estimated ± 13.5 pp, low confidence |
| DeepsecBench-AA | 4.9% | estimated ± 9.0 pp, low confidence |
| AA ITBench | 39.7% | estimated ± 4.1 pp, high confidence |
| DeepSWE | 24.3% | estimated ± 8.4 pp, low confidence |
| DRACO | 65.9% | estimated ± 2.9 pp, low confidence |
| Toolathlon-Verified | 51.5% | estimated ± 7.4 pp, medium confidence |
| Claw-Eval | 62.7% | estimated ± 6.4 pp, medium confidence |
| DeepPlanning | 26.9% | estimated ± 9.1 pp, low confidence |
| MCP Atlas | 54.0% | estimated ± 10.3 pp, low confidence |
| QwenClawBench | 52.0% | estimated ± 2.0 pp, medium confidence |
| Toolathlon | 37.2% | estimated ± 4.2 pp, high confidence |
| ResearchClawBench | 15.8% | estimated ± 2.3 pp, medium confidence |
| Agents' Last Exam | 16.8% | estimated ± 11.6 pp, low confidence |
| LifeVal | 51.4% | estimated ± 1.6 pp, medium confidence |
| PinchBench | 89.4% | estimated ± 3.6 pp, medium confidence |